188宝金博页面版

Routing Stability in Static Wireless Mesh NetworksKrishna Ramachandran, Irfan Sheriff, Elizabeth Belding, Kevin AlmerothUniversity of California, Santa BarbaraAbstract. Considerable research has focused on the design of routing protocolsfor wireless multi-hop networks. Yet, very little is understood about the stabil-ity of routes in such networks. This understanding is important in the design ofwireless routing protocols, and in network planning and management. In this pa-per, we present results from our measurement-based characterization of routingstability in two network deployments, the UCSB MeshNet and the MIT Roofnet.To conduct these case studies, we collected detailed link quality information overseveral days from each ofthese networks. Using this information, we investigatedrouting stability in terms ofroute-level characteristics, such as prevalence, persis-tence and flapping. Our key findings are the following: wireless multi-hop routesare weakly dominated by a single route; dominant routes are extremely short-lived due to excessive route flapping; and simple stabilization techniques, such ashysteresis thresholds, can provide a significant improvement in route persistence.1IntroductionApplications, such as ‘last-mile’ Internet delivery, public safety, and distributed sens-ing, are driving the deployment oflarge-scale multi-hop wireless networks, also knownas mesh networks. Wireless routers in such networks are typically stationary. Notabledeployments of mesh networks in the United States include those in Mountain Viewand Philadelphia.Routes in these networks are expected to be unstable. One reason is that wirelesslinks vary widely in their qualities because of multi-path fading effects, external inter-ference and weather conditions. Link quality fluctuations can lead to variations in thequality ofmesh routes, which can result in route fluctuations. This degree ofinstabilityis unique to wireless networks.Current routing protocols are not intelligent enough to consider routing stabilityduring the selection ofroutes. A majority ofthe routing protocols [7] [12] [17] will ig-nore the fact that a route initially discovered has become sub-optimal over time. Routerecomputation is typically triggered by only route breaks and route timeouts. This ap-proach can severely degrade network performance.Other routing protocols [3][9] periodically re-evaluate the quality of a route. Theevaluation periodicity depends on the rate at which routing protocol control messagesare exchanged. This approach fails to adapt to route quality variations that occur atsmaller time-scales. However, by always picking the best route available, the resultingrouting instability can lead to routing pathologies, such as packet reordering [4], whichcan severely degrade network performance.An investigation of routing stability in mesh networks is clearly important. We re-quire a routing protocol that provides the best tradeoffbetween performance adaptabil-ity and routing stability. A comprehensive understanding of routing stability vis-a-visrouting performance can help us design more efficient routing protocols.